Comparing the Effectiveness of Error-Correction Strategies in Discrete Trial Training

Abstract

Error-correction strategies are essential considerations for behavior analysts implementing discrete trial training with children with autism. The research literature, however, is still lacking in the number of studies that compare and evaluate error-correction procedures. The purpose of this study was to compare two error-correction strategies: Independent Probe and Delay across learners with autism in an intensive intervention program. Two studies were conducted. The first study compared the two procedures across receptive tasks for 3 individuals, and differential effects were seen across learners. The second study compared the two procedures across tact trials with two of the same learners and found that individual differences were noted, but in addition, the more effective error-correction strategy was consistent across the two verbal operants (i.e., receptive in Study 1, tacts in Study 2). These combined studies suggest the effectiveness of error-correction strategies may be individualized to the learner but may generalize across operants.

An assessment of error-correction procedures for learners with autism

“…These findings are consistent with other studies showing the effectiveness of procedures that require multiple responses from students following errors, a strategy that is often called positive practice or directed rehearsal (e.g., Foxx & Jones, 1978, Ollendick, Matson, Esveldt-Dawson, & Shapiro, 1980. Turan et al (2012), who also obtained inconsistent results across subjects when comparing two commonly recommended error-correction procedures for discrete-trial teaching, suggested that instruction should be individualized across learners. Smith et al (2006) compared three conditions that involved providing a different response following errors: (a) no consequences, (b) vocal feedback only (saying "no"), and (c) a model of the correct response.…”

“…All six students acquired more words assigned to the ASR condition than to the NR condition. Results of other studies on different errorcorrection procedures have been less consistent (Rodgers & Iwata, 1991;Smith et al, 2006;Turan, Moroz, & Croteau, 2012). Results of a subsequent evaluation, in which the experimenter required subjects to repeat either the targeted word or a nontargeted word contingent on errors, indicated that the multiple-response repetition error-correction procedure functioned as punishment for incorrect responding.…”

“…Nonetheless, one subject performed better with vocal feedback alone, and one subject performed similarly across all conditions. Turan et al (2012), who also obtained inconsistent results across subjects when comparing two commonly recommended error-correction procedures for discrete-trial teaching, suggested that instruction should be individualized across learners.…”

Inserting mastered targets during error correction when teaching skills to children with autism

“…Some authors have suggested that therapists should insert a known (i.e., previously acquired) target between a prompted response and an opportunity for an independent response when they implement error‐correction procedures (e.g., Bondy & Frost, ; Turan, Moroz, & Croteau, ). However, in most studies on this error‐correction method, the therapist delivered a prompt after an incorrect response and then immediately set up the occasion for the individual to respond independently (e.g., by delivering the S D again with no prompt).…”

The effects of error correction with and without reinforcement on skill acquisition and preferences of children with autism spectrum disorder

“…A variety of skills were targeted in these three studies, including requesting through picture selection (Carr & Felce, 2008), sight-word reading (Marvin et al, 2010), receptive selection, and labeling (Turan et al, 2012). The error-correction procedures used in the studies varied significantly.…”